[Feature Request]: Is it possible to edit the model's system prompt words when creating an agent?

[SoftCharacter]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (Language Policy).
• Non-english title submitions will be closed directly ( 非英文标题的提交将会被直接关闭 ) (Language Policy).
• Please do not modify this template :) and fill in all the required fields.

Is your feature request related to a problem?

When I was creating a websearch agent, I found that calling the model's rewrite node and keyword node returned data that included not only the result data, but also what the model was thinking about. This results in unwanted errors in subsequent search nodes and crawler nodes

Describe the feature you'd like

I'd like to be able to edit the model's system prompt word when the agent is created, instead of only being able to edit him when generating the answer, as is currently the case. Or would like to be able to add a node for parsing strings (where the user can write regular expressions for example)

Describe implementation you've considered

No response

Documentation, adoption, use case

Additional information

No response

[ZhenhangTung]

@SoftCharacter Thanks for sharing this!

I recommend using the Generate component for your case—it gives you more control over the model output and helps avoid unintended content leaking into downstream nodes.

We’re also in the process of designing the Agent component, where you’ll be able to fully customize both System and User prompts directly in the configuration.

Appreciate your feedback—it aligns well with what we’re building!